Criterion A: Problem specification
Criterion A is the starting point of the whole solution. It says what problem you are solving, for whom, how you will know you have solved it, and why the technology you chose suits it. The IB guide says the success criteria from A are used again in planning (B), development (D) and evaluation (E), and the IA Guide adds that the testing strategy in C should line up with them too.
What you need to produce
- A description of the problem scenario with measurable requirements
- An explained choice of computational context (language, platform, libraries, key concepts)
- A list of success criteria, each with a code (SC1, SC2…); the IA Guide recommends 8–10
Word budget
About 200 counted words: roughly 110 for the scenario, 70 for the context and 20 for the sentence that introduces the success criteria. The IB FAQ (November 2026) says success criteria written as bullet points are not counted, so list them as bullets, one short, testable statement each. See how words are counted and allocated.
Subsections
| Subsection | Words (plan) | Suggested range | Main evidence |
|---|---|---|---|
| A1 Problem scenario | 110 | 95–120 | prose |
| A2 Computational context | 70 | 60–80 | prose |
| A3 Success criteria | 20 | 5–30 | SC1–SC10 as bullet points, not counted |
Each subsection below has the same seven parts: purpose, what to include, suggested length, a worked example, supporting evidence, lessons from the IB exemplars and a completion checklist. Labels show where each piece of guidance comes from:
Official level descriptors for Criterion A (best-fit marking)
| Marks | Level descriptor (IB Computer Science guide) |
|---|---|
| 0 | The response does not reach a standard described by the descriptors below. |
| 1–2 | outlines a problem scenario states limited success criteria outlines the nature of the solution in a computational context |
| 3–4 | describes the problem scenario in terms of its measurable solution requirements states appropriate success criteria explains the choice of computational context for the solution |
1 Purpose
Establish a real, specific problem and the measurable requirements a solution must meet, so that an examiner can see exactly what “solved” would mean before any software is described.
2 What to include
- IB requirementDescribe the problem scenario in terms of its measurable solution requirements (3–4 band). It can come from the world around you, another subject or a current issue in computing.
- IA GuideBase it on a real person and a real situation, not “for fun” or “to test my skills”.
- IA GuideSay clearly who is affected.
- IA GuideDescribe the current process and why it fails, with figures where you can (how often, how long, how many).
- IA GuideInclude at least one measurable requirement; aim for several.
- Teaching tipFinish with the requirements in one sentence or two. Each one should later become one or more success criteria, so the link between A1 and A3 is obvious.
- IB requirementCheck you have the technical skills, hardware, software and access to data needed to solve it.
3 Suggested length
About 95–120 counted words (planning allocation 110). Prose is the right format here: the examiner needs to follow the story of the problem. A small table of observations (for example, a tally of double bookings) is fine as evidence but should not replace the explanation.
4 Worked example
Our school's six music practice rooms (P1–P6) are shared by about 140 students and booked on a paper sheet outside the music office. Over two weeks I observed the sheet and spoke to the head of music, Ms Okafor. I found 11 double-booked slots, costing her about 20 minutes a week in disputes; 14 of 42 lunchtime bookings unused; and drummers in neighbouring rooms making both rooms unusable.
A solution must prevent overlapping bookings, limit each student to three bookings a week, release rooms not checked in within 10 minutes, stop loud bookings next to each other, and let students book from a phone in under a minute.
5 Supporting evidence (to be added to your appendix)
- Evidence you gathered yourself: a photo of the current paper process (with permission), a tally table, a short summary of a conversation with the person affected, with their consent.
- Keep raw evidence (interview notes, photos) in the appendix and refer to it; see the ethics guidance on the IA guide page.
6 Lessons from the IB exemplars
From the IA Documentation Guide's analysis of real, examiner-graded IB Computer Science IAs.
- A scenario with zero measurable requirements caps this criterion, whatever else is strong.For your IA: Put numbers and testable limits in the scenario itself (under a minute, 3 per week, 10 minutes), not only in the success criteria.
7 Completion checklist
Ticks are saved in this browser only, so you can use the list while you draft.
1 Purpose
Explain which area of computing your solution uses and why those choices fit this particular problem, so that the examiner sees a reasoned decision rather than a default.
2 What to include
- IB requirementExplain the choice of computational context (3–4 band); “outline” only reaches 1–2. The IB defines the computational context as “the specific area of computing that is selected to be used in the solution”.
- IA GuideName your language, libraries and platform explicitly.
- IA GuideJustify each choice for this problem. Avoid generic reasons such as “Python is easy”.
- IA GuideMention any core technical concept your domain needs, for example networking, encryption, concurrency or machine learning, and say why it is needed.
- Teaching tipA good test: if your justification would be equally true for any project, it is too generic. Tie it to a requirement from A1 (phones, no installs, two users at once).
3 Suggested length
About 60–80 counted words (planning allocation 70). Prose, because the marks are for explaining. Do not hide the reasoning in a table of pros and cons.
4 Worked example
PracticePal will be a Python Flask web app with an SQLite database. A web app suits students' phones and school laptops, which block installations. Flask is small, so I write every route and rule myself. SQLite is one file the IT technician can host on the intranet server. Two concepts are central: concurrency, because two students may book one slot at once, and password hashing.
5 Supporting evidence (to be added to your appendix)
- Usually none needed beyond the text. If a hosting constraint drives your choice (for example, the school server only runs Python), you can refer to written confirmation in the appendix.
- Related site pages: servers, application protocols such as HTTP, transactions.
6 Lessons from the IB exemplars
From the IA Documentation Guide's analysis of real, examiner-graded IB Computer Science IAs.
- Generic tech justification (“Python is easy”) cost marks even in an otherwise 4/4 paper.For your IA: Give a reason that only makes sense for your problem, and link it to a requirement in A1.
7 Completion checklist
Ticks are saved in this browser only, so you can use the list while you draft.
1 Purpose
Turn the requirements into a list of specific, testable outcomes. These are the measuring stick for the rest of the IA: they appear again in the plan, the test plan, the development evidence and the evaluation.
2 What to include
- IB requirementState appropriate success criteria: measurable outcomes derived from the solution requirements that indicate the successful development of the product.
- IA GuideAim for 8–10 specific, testable, measurable statements.
- IB FAQWrite them as bullet points. The top descriptor only asks you to “state” success criteria, so no extended writing is needed: bulleted criteria are not counted in the 2,000 words, but criteria written as paragraphs of extended text are.
- IA GuideGive each one a code: SC1, SC2, SC3… List them once, here. In B, C, D and E refer to them by code only, never repeating their text, so your later paragraphs stay short.
- IA GuideOnly list what you will actually build.
- Teaching tipWrite each one so a stranger could test it: a user, an action, a condition and a measurable result. Avoid “looks nice”, “runs well” or “is user-friendly”.
- Teaching tipCover the whole decomposition: every major component should be checked by at least one criterion, including outputs such as reports or notifications.
3 Suggested length
About 5–30 counted words (planning allocation 20), for any sentence around the list. The bulleted list itself is not counted (IB FAQ), as long as each criterion is a brief statement, not a paragraph. Keep each one to a single testable statement, and write them once: after A, use only the codes.
4 Worked example
- SC1 Users log in by school email and hashed password; only the head of music opens admin screens.
- SC2 A grid shows each room's 30-minute slots, 07:30–17:30 Monday–Friday, this week and next.
- SC3 A free slot is booked in at most three taps, confirmed within 2 seconds.
- SC4 Overlapping bookings for one room are never stored, even if submitted simultaneously.
- SC5 A student's fourth booking in a Monday–Sunday week is refused with a reason.
- SC6 A loud booking is refused with a reason if a neighbouring room has an overlapping loud booking.
- SC7 A freed slot is offered to the first student waiting, who has 15 minutes to accept.
- SC8 A booking not checked in within 10 minutes of its start becomes a no-show and is released.
- SC9 The head of music can block a room for dates, cancelling affected bookings and emailing students.
- SC10 The head of music can download a weekly CSV per room: bookings, percentage used, no-shows.
From here on, each criterion is referred to only by its code.
5 Supporting evidence (to be added to your appendix)
- The list is the evidence. Keep the codes stable: if you remove a criterion after your teacher sees the draft, renumber everywhere.
- Trace each criterion forwards: plan task (B), test (C), evidence (D), evaluation row (E). The worked examples on this site use SC1–SC10 in exactly that way.
6 Lessons from the IB exemplars
From the IA Documentation Guide's analysis of real, examiner-graded IB Computer Science IAs.
- Projects with only 5–6 success criteria scored the two lowest totals in the whole set (16/30 and 13/30).For your IA: Aim for 8–10. Too few criteria leaves the plan, tests and evaluation with too little to work with.
- The guide's checklist warns that fewer than 8 success criteria drags down every later criterion, not just A.For your IA: Count yours now. If you have fewer than 8, look for components of your decomposition with no criterion.
7 Completion checklist
Ticks are saved in this browser only, so you can use the list while you draft.
Criterion A worked examples: word total
| Subsection | Planning allocation | Worked example (counted words) |
|---|---|---|
| A1 Problem scenario | 110 | 109 |
| A2 Computational context | 70 | 65 |
| A3 Success criteria | 20 | 12 |
| Total | 200 | 186 |
The fictional PracticePal documentation uses 186 of the 200 counted words planned for Criterion A. Across all five criteria it stays within 2,000; see the budget table.